Archives of the TeradataForum
Message Posted: Wed, 10 Jun 2015 @ 10:30:44 GMT
| Subj: | | Re: Write Lock - Partition level |
| |
| From: | | Sravan.Bodla |
Anomy.Anom wrote:
| | You probably meant to say: | |
| | In DEV they were updating a non PI column which was placing table locks (so updates took turns but succeeded) and in PROD the updates were
on PI column which is a RowHash level lock leading to Deadlock (due to the application design holding some locks already and requesting
others). | |
Nope. It's in reverse.
In DEV they were updating a non PI column which was placing ROW HASH locks (so updates succeeded) and in PROD the updates were on PI column
Placing table level lock, leading to Deadlock (due to the application design/huge parallel sessions, some sessions holding locks already and
others requesting/waiting for access).
Hope this is clear now! :-)
Thanks!
Sravan
| 
